Download Drug Monitoring: Simultaneous Analysis of Lamotrigine

Journal of Chromatographic Science, Vol. 45, October 2007
Drug Monitoring: Simultaneous Analysis of Lamotrigine,
Oxcarbazepine, 10-Hydroxycarbazepine, and
Zonisamide by HPLC–UV and a Rapid GC Method Using
a Nitrogen-Phosphorus Detector for Levetiracetam
Elizabeth Greiner-Sosanko, Spiros Giannoutsos, Darla R. Lower, Mohamed A. Virji, and Matthew D. Krasowski*
Toxicology and Therapeutic Drug Monitoring Laboratory, Division of Clinical Chemistry, Department of Pathology, University of Pittsburgh
Medical Center, Children’s Hospital Main Tower 5812, Pittsburgh, PA 15213
Abstract
A high-performance liquid chromatography (HPLC) assay using UV
detection is described for the simultaneous measurement of the
newer generation anti-epileptic medications lamotrigine,
oxcarbazepine (parent drug and active metabolite 10hydroxycarbazepine), and zonisamide. Detection of all four
compounds can be done at 230 nm; however, there is a potential
interference with zonisamide in patients on clonazepam therapy.
Therefore, the method uses dual wavelength detection: 230 nm for
oxcarbazepine and 10-hydroxycarbazepine and 270 nm for
lamotrigine and zonisamide. In addition, a simple gas
chromatography method using a nitrogen-phosphorus detector is
described for the measurement of levetiracetam, another of the
recently approved anti-epileptic medications. For both methods,
limits of quantitation, linearities, accuracies, and imprecisions
cover the therapeutic range for drug monitoring of patients. A wide
variety of clinical drugs, including other anti-epileptic drugs, do not
interfere with these assays. These procedures would be of special
interest to clinical laboratories, particularly due to the limited
availability of immunoassays for newer generation anti-epileptic
medications and that therapeutic uses of these drugs are expanding
beyond epilepsy to other neurologic and psychiatric disorders.
immunoassays for the newer anti-convulsants, if commercially
available, tend to run on a limited number of specialized platforms and not on common high throughput chemistry analyzers. This necessitates the development of robust and
straightforward chromatographic assays for the new antiepileptic agents. This article reports methods appropriate for
drug monitoring of four of the most frequently prescribed of the
newer anti-epileptic medications: lamotrigine, levetiracetam,
oxcarbazepine, and zonisamide. The chemical structures of these
four drugs, along with the two major metabolites of oxcarbazepine, are shown in Figure 1.
Lamotrigine (Lamictal, GlaxoSmithKline) is a broad-spectrum anti-epileptic drug of the phenyltriazine class chemically
unrelated to other anti-convulsants that was approved by the
Food and Drug Administration (FDA) for treatment of epilepsy in
1994 (1,2). In 2003, lamotrigine received an additional approval
for treatment of bipolar disorder. Lamotrigine is predominantly
Introduction
During the past twelve years, a number of novel anti-epileptic
agents have been introduced into clinical practice in the United
States, considerably expanding the therapeutic options beyond
traditional medications such as carbamazepine, phenobarbital,
phenytoin, primidone, and valproic acid. For therapeutic drug
monitoring, immunoassays for the traditional anti-epileptic
drugs have been available for years from multiple companies, in
many cases on high-throughput chemistry analyzers. In contrast, development of similar assays for the newer anti-epileptic
agents has been slow. Unlike the traditional anti-convulsants,
* Author to whom correspondence should be addressed: email [email protected].
616
Figure 1. Chemical structures of lamotrigine, levetiracetam, zonisamide,
oxcarbazepine, 10-hydroxycarbazepine (pharmacologically active metabolite of oxcarbazepine), and 10,11-dihydroxy-trans-10,11-dihydrocarbamazepine (therapeutically inactive metabolite of oxcarbazepine).
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher’s permission.
Journal of Chromatographic Science, Vol. 45, October 2007
metabolized in the liver by glucuronidation (3). Lamotrigine has
an average elimination half-life of 33 h, although this can be
influenced by concomitant therapy with other medications that
inhibit or induce expression of hepatic drug-metabolizing
enzymes. Therapeutic drug monitoring of lamotrigine targets
the parent drug as little clinical benefit has been found for monitoring the glucuronide metabolite (3). Several methods have
been used for determination of lamotrigine concentrations
including high-performance liquid chromatography (HPLC)
(4–8), gas chromatography (GC) with a nitrogen-phosphorus
detector (NPD) (9), and radioimmunoassay (10).
Levetiracetam (Keppra, UCB Pharma) is an anti-epileptic drug
licensed for the treatment of partial seizures in adults and is
sometimes used in combination with other anti-epileptic drugs.
It is also effective in children with partial seizures and may also
be useful in select patients with generalized seizures (11). The
mechanism of action of levetiracetam is unclear but may involve
inhibition of voltage-activated calcium channels in the brain.
The elimination half-life of levetiracetam is approximately 6 to 8
h in adults. Approximately 66% of the drug is excreted
unchanged and 27% as an inactive hydrolysis product.
Levetiracetam is predominantly cleared by the kidneys so that
elimination parallels kidney function (3,12). Levetiracetam is
now available in an intravenous formulation in both Europe and
the United States, allowing this drug to be used more frequently
in urgent clinical situations such as status epilepticus, where
faster turnaround time on levetiracetam plasma concentrations
may be needed clinically. A number of methods have been developed for determination of levetiracetam concentrations
including HPLC (13–17), gas chromatography (18) (note that
little experimental detail was provided in this report as analysis
was done by a contract research organization), GC–mass spectrometry (MS) (19), and liquid chromatography–MS–MS (20).
Levetiracetam raises some challenges with extraction and HPLC
analysis due to the compound’s high polarity and lack of a good
chromophore for UV absorption (14,15). HPLC analysis of levetiracetam, thus, requires either derivatization or use of shorter
wavelength UV light that other compounds absorb non-specificity.
Oxcarbazepine (Trileptal, Novartis) is the 10-keto analog of
carbamazepine with similar therapeutic effects but with fewer
adverse effects, including less ability to induce the expression of
hepatic drug-metabolizing enzymes (1,2). Oxcarbazepine is now
a first-line drug for the management of partial seizures. Like carbamazepine, oxcarbazepine blocks voltage-gated sodium channels. Following oral administration and absorption,
oxcarbazepine is rapidly reduced by cytosolic arylketone reductases to its active metabolite 10-hydroxy-10,11-dihydrocarbamazepine (10-hydroxycarbazepine). 10-Hydroxycarbazepine
accounts for the majority of anti-epileptic activity following
oxcarbazepine administration, and, therefore, therapeutic drug
monitoring generally targets this metabolite (3,21). 10Hydroxycarbazepine is eliminated by conjugation with glucuronic acid. Minor amounts (4% of the ingested dose of
oxcarbazepine) are further oxidized from 10-hydroxycarbazepine
to 10,11-dihydroxy-trans-10,11-dihydrocarbamazepine, which is
pharmacologically inactive with respect to treating seizures (21).
A number of methods have been used to determine oxcar-
bazepine and 10-hydroxycarbazepine metabolite concentrations
including HPLC (8,22–28), GC (29), and micellar electrokinetic
chromatography (30).
Zonisamide (Zonegran, Eisai Co.) is an anti-epileptic drug
licensed in the United States for the treatment of partial seizures
in adults (1,2). Zonisamide is also used in Lennox-Gestaut syndrome (often a treatment-refractory type of epilepsy), West’s syndrome (a form of infantile epilepsy), bipolar disorder, and
migraine headaches. Zonisamide has an elimination half-life of
approximately 50–70 h in adults and is extensively metabolized
by acetylation and conjugation. Zonisamide is a substrate of
cytochrome P450 2C19 and 3A4 and is subject to drug–drug
interactions with inhibitors or inducers of these liver enzymes
(3). Therapeutic drug monitoring of zonisamide has involved
determination of serum/plasma concentrations of only the
parent compound, as no advantage has been found in monitoring metabolite concentrations (3). Several HPLC methods
have been reported for zonisamide (14,31,32).
An HPLC method that simultaneously measures lamotrigine,
oxcarbazepine, 10-hydroxycarbazepine, and zonisamide concentrations in plasma/serum has not been reported. These medications are increasingly being used in settings outside of epilepsy
such as bipolar disorder and migraine headaches and sometimes
in combination for refractory epilepsy or other conditions where
serum/plasma levels of multiple drugs are needed simultaneously. Therefore, having an assay that can analyze multiple drugs
simultaneously with a single extraction is very helpful in
increasing throughput and limiting manual steps. In this paper,
an HPLC–UV assay is described that accomplishes this goal. In
addition, a GC assay using a NPD for levetiracetam is described
that can support rapid turnaround of levetiracetam plasma concentrations. This assay uses a methylene chloride-based extraction that achieves excellent specificity.
Experimental
Reagents
Lamotrigine was generously supplied by GlaxoSmithKline
(Triangle Park, NC). 10-Hydroxycarbazepine and 10,11-dihydroxy-trans-10,11-dihydrocarbamazepine were kindly supplied
by Novartis (Basel, Switzerland). Levetiracetam reference standard was generously supplied by UCB Pharma (Smyrna, GA).
Oxcarbazepine, zonisamide, and chloramphenicol were obtained
from Sigma-Aldrich (St. Louis, MO). Mepivacaine (1 mg/mL in
methanol) was purchased from Alltech (Lancaster, PA).
Acetonitrile, methanol, and water were HPLC grade (Fisher
Scientific, Pittsburgh, PA). Ethylacetate, potassium phosphate,
methylene chloride, and phosphoric acid were analytical grade
(Fisher). Drug-free human plasma was obtained from the
University of Pittsburgh Medical Center Central Blood Bank.
HPLC analysis of lamotrigine, oxcarbazepine,
10-hydroxycarbazepine, and zonisamide
The HPLC system consisted of a Waters 717 Plus Autosampler
and Systems Controller and a Waters 2487 Dual Absorbance
Detector (Milford, MA). The separation was performed at 22°C
617
Journal of Chromatographic Science, Vol. 45, October 2007
with a Supelcosil LC-18 column (25 cm × 4.6 mm internal diameter, 5 µM silica particles; Supelco, Bellefonte, PA). The mobile
phase was a mixture of aqueous 3mM potassium phosphate
buffer (adjusted to pH 3.7 with 5% phosphoric acid) and acetonitrile (65:35) at a flow rate of 1.2 mL/min. Oxcarbazepine and 10hydroxycarbazepine are monitored at 230 nm while lamotrigine
and zonisamide are monitored at 270 nm (all other HPLC conditions are identical for the four compounds). Stock solutions of
oxcarbazepine, 10-hydroxycarbazepine, lamotrigine, zonisamide, and the internal standard chloramphenicol were at
1000 µg/mL in HPLC grade methanol. To 250 µL of sample (calibrator, control, or patient sample) in a 16 × 125 mm glass culture tube, 100 µL of chloramphenicol solution, 1.5 mL 10 mM
NaOH, and 4.0 mL ethylacetate were added. The samples were
immediately vortexed for 1 min and centrifuged at 1,700 × g for
5 min. The organic (upper) layer was transferred to a conical
tube and dried completely at 40°C using a nitrogen evaporator.
The sample was then reconstituted with 100 µL mobile phase,
vortexed, and transferred to an autosampler vial for HPLC analysis. A range of possible interfering substances were evaluated by
extracting and analyzing Liquichek Immunoassay Plus Control
(Bio-Rad, Irvine, CA). This serum product has over 90 compounds, including therapeutic drugs (e.g., acetaminophen,
amikacin, amiodarone, amitriptyline, caffeine, carbamazepine,
cyclosporine, desipramine, digoxin, disopyramide, flecainide,
gentamicin, ibuprofen, imipramine, lidocaine, lithium,
netilmycin, nortriptyline, propanolol, quinidine, salicylate,
tobramycin, and valproic acid), vitamins, and steroid hormones.
Other compounds not found in the Liquichek product were also
screened for interference, particularly neurologic and psychiatric medications such as baclofen, clonazepam, felbamate,
tiagabine, and topiramate.
GC analysis of levetiracetam
The GC system was a Perkin-Elmer Autosystem XL GC with
NPD and a HP-5MS capillary column (5% phenyl methyl
siloxane, 30 m × 0.251 mm × 0.25 µM; J&W Scientific, Folsom,
CA). The internal standard mepivacaine solution was prepared at
10 µg/mL in methanol. Several different possible extraction procedures were evaluated before determining that extraction with
methylene chloride provided excellent specificity and recovery.
Extraction with ethylacetate would be an alternative but leads to
less specificity. To 250 µL of sample (calibrator, control, or
patient samples) in a 16 × 125 mm screw-top culture tube, 50 µL
of mepivacaine solution, 250 µL of phosphate buffer (290 mM
KH2PO4, 210 mM Na2HPO4, adjusted to pH 6.4 with 10% phosphoric acid), and 5 mL methylene chloride were added. The samples were immediately vortexed for 1 min and centrifuged at
2,400 rpm for 5 min. The bottom organic layer was then transferred to a 10 mL conical centrifuge tube and evaporated to dryness at 40°C using a nitrogen evaporator. The sample was then
reconstituted with 50 µL methanol, vortexed, and transferred to
an autosampler vial for GC analysis. The oven temperature began
at 90°C, increased by 42.0°C/min to 190°C, and then increased
by 10.0°C/min to 250°C. Total analysis time was 8 min. Different
temperatures and ramping speeds were extensively evaluated.
Attempts to shorten the GC run-time by increasing the ramp
speed degraded the quality of the chromatograms. Levetiracetam
618
is detected poorly by a flame ionization detector. Our initial
attempts to use a flame ionization detector showed sensitivity
that was at least ten-fold less than that with a NPD. The limit of
quantitation using a flame ionization detector is estimated at
approximately 10 µg/mL, insufficiently low to capture the therapeutically meaningful range of serum/plasma concentrations.
Interfering substances were assessed as described for the HPLC
method described.
Determination of recovery, intra-day and inter-day
precision, accuracy, linearity, and limit of detection and
quantitation
Intraday precision and accuracy were evaluated by the analysis
of spiked samples. The precision and accuracy for interday comparisons were assessed at the same concentration and summarized as coefficient of variation (CV%) and relative deviation
(RD%), respectively. Linearity of the methods were assessed by
analyzing drug-free human plasma to which analytes were
added. Calibrators were sent to a reference laboratory to validate
concentrations. The linear range was considered to be the concentration range for which the calculated concentrations were
within 85–115% of the spiked concentration. Limit of detection
was determined using the signal-to-noise of 3:1 by comparing
test results from samples with known concentrations of analytes
with drug-free plasma (blank) samples. The limit of quantitation
was defined as the lowest drug concentration that can be determined reproducibly (coefficient of variation < 20%) and accurately (percent error < 20%).
Results and Discussion
HPLC assays for lamotrigine, oxcarbazepine,
10-hydroxycarbazepine, and zonisamide
Lamotrigine absorbs UV light well in the 270–306 nm range,
and a number of HPLC methods utilize wavelengths within this
range for detection of lamotrigine (7,33). Fewer methods
describe HPLC detection of zonisamide, but this compounds
optimally absorbs UV light at a somewhat lower UV range
(32,34). Oxcarbazepine and 10-hydroxycarbazepine absorb
poorly at 270 nm but better absorption occurs at 215–240 nm
(22,25,27).
Under the described extraction and HPLC run conditions,
lamotrigine, oxcarbazepine, 10-hydroxycarbazepine, and zonisamide could all be adequately detected at 230 nm (Figure 2A).
However, during method development, substantial interference
with the zonisamide peak was noted at 230 nm wavelength
detection in several patients who were also receiving clonazepam
therapy. Testing of blank plasma spiked with clonazepam showed
that the parent drug did not account for the interference. We
speculate that clonazepam metabolites likely contributed to the
interference in these patients, but there is a lack of standards of
these metabolites to further test this hypothesis.
Therefore, we searched for two wavelengths that could be used
for detection, while still keeping the same extraction and HPLC
conditions. A solution was found with 230 nm for oxcarbazepine
and 10-hydroxycarbazepine and 270 nm for lamotrigine and zon-
Journal of Chromatographic Science, Vol. 45, October 2007
AU
270 nm, see previously for description of interference at 230
nm), epoxycarbamazepine, ethosuximide, felbamate,
gabapentin, levetiracetam, phenobarbital, phenytoin, primidone,
tiagabine, topiramate, and valproic acid. None of the therapeutic
drugs tested exhibited any interference. In addition, no peak
interferences were found in any of the batches of drug-free
plasma. Separate experiments with the reference standard of the
dihydroxy metabolite of oxcarbazepine (10,11-dihydroxy-trans10,11-dihydrocarbamazepine) showed that this oxcarbazepine
metabolite did not interfere with the HPLC analysis. As described
in the introduction, this metabolite is pharmacologically inactive with respect to epilepsy therapy, rarely detected in patients
Time (min)
AU
AU
isamide. Using this method, the retention times of 10-hydroxycarbazepine, zonisamide, chloramphenicol (internal standard),
oxcarbazepine, and lamotrigine were 3.5, 4.5, 4.9, 5.2, and 5.9
min, respectively (Figures 2A and 2B), using drug-free plasma
spiked with pure compounds. An example of a chromatogram
from the plasma of a patient chronically taking both lamotrigine
and zonisamide is shown in Figure 3A (using 270 nm wavelength
detection). An example of a chromatogram from the plasma of a
patient chronically taking oxcarbazepine is shown in Figure 3B
(using 230 nm wavelength detection); in this example, most of
the parent oxcarbazepine had been converted to 10-hydroxycarbazepine in this patient, as is typical in samples received for therapeutic drug monitoring. The use of a dual-wavelength detector
allows for rapid simultaneous analysis of 10-hydroxycarbazepine,
zonisamide, chloramphenicol (internal standard), oxcarbazepine, and lamotrigine. Alternatively, for laboratories that
have a single variable wavelength detector, the single extraction
can still be used with separate injections and analyses at 230 nm
(10-hydroxycarbazepine and oxcarbazepine) and 270 nm (lamotrigine and zonisamide).
A wide variety of therapeutic drugs were tested for interference
at both 230 and 270 nm, including other anti-epileptic and neurologic medications (and in some cases their metabolites) such
as baclofen, carbamazepine, clonazepam (no interference at
Time (min)
Figure 3. Chromatogram of a plasma sample from a patient on chronic oxcarbazepine therapy using 230 nm wavelength detection. Determined concentrations: 10-hydroxycarbazepine (11.2 µg/mL) and oxcarbazepine (< 0.75
µg/mL) (A). Chromatogram of a plasma sample from a patient on chronic lamotrigine and zonisamide therapy using 270 nm wavelength detection (B).
Determined concentrations: lamotrigine (4.3 µg/mL) and zonisamide (16.5
µg/mL).
Time (min)
Figure 2. Chromatogram from HPLC analysis of a blank plasma sample
spiked with lamotrigine (5 µg/mL), zonisamide (15 µg/mL), oxcarbazepine (2
µg/mL), 10-hydroxycarbazepine (15 µg/mL) and the internal standard chloramphenicol using 230 nm wavelength detection. Note that all compounds
are readily detected at this wavelength but potential interference with zonisamide occurs in patients taking clonazepam (not shown) (A).
Chromatogram from HPLC analysis of a blank plasma sample spiked with
lamotrigine (5 µg/mL), zonisamide (15 µg/mL), oxcarbazepine (2 µg/mL), 10hydroxycarbazepine (15 µg/mL) and the internal standard chloramphenicol
using 270 nm wavelength detection (B). At this wavelength, 10-hydroxycarbazepine (*) and oxcarbazepine (**) are poorly detected compared to 230
nm.
Response (mV)
AU
Time (min)
Time (min)
Figure 4. Chromatogram from GC analysis of a plasma sample from a patient
on chronic levetiracetam therapy. Determined levetiracetam concentration:
18.8 µg/mL.
619
Journal of Chromatographic Science, Vol. 45, October 2007
on oxcarbazepine therapy, and not of interest for therapeutic
drug monitoring purposes.
The method exhibited a good linearity over a concentration
range of 1–30 µg/mL for lamotrigine, 1–30 µg/mL for oxcarbazepine, 1–40 µg/mL for 10-hydroxycarbazepine, and 1–40
µg/mL for zonisamide. This covers the therapeutic range for
these compounds. For lamotrigine, the linear regression equation for comparison of determined versus spiked concentration
was: y = 1.0x + 0.40, with r = 0.998 and a standard error value of
0.62. For 10-hydroxycarbazepine, the linear regression equation
for comparison of determined versus spiked concentration was: y = 0.997x – 0.07, with r = 0.997
Table I. Intra- and Interday Accuracy and Precision for the HPLC Assay for
and a standard error value of 0.99. For oxcarDetetermination of Lamotrigine, Zonisamide, Oxcarbazepine, and 10bazepine, the linear regression equation for
Hydroxycarbazepine Concentrations in Human Plasma/Serum (All
comparison of determined versus spiked concenConcentrations in g/mL)
tration was: y = 0.998x + 0.14 with r = 0.996 and
a standard error value of 1.23. For zonisamide,
Lamotrigine
the linear regression equation for comparison of
Intraday (n = 5)
determined versus spiked concentration was: y =
Nominal concentration
5
12
20
0.998x + 0.19, with r = 0.998 and a standard
Mean
5.22 ± 0.23*
12.32 ± 0.61
12.32 ± 0.61
error value of 1.36.
Accuracy as RD (%)
4.4
2.6
1.3
Precision as CV (%)
4.4
4.9
5.5
Results of intra- and interday accuracy and
precision are shown in Table I. The lower limit of
Interday (n = 12)
detection was 0.15 µg/mL for lamotrigine, 0.25
Nominal concentration
7
12
µg/mL for oxcarbazepine, 0.15 µg/mL for 10Mean
7.28 ± 0.35
12.10 ± 0.76
hydroxycarbazepine, and 0.75 µg/mL for zonAccuracy as RD (%)
4.0
0.8
isamide. The limit of quantitation was 0.5 µg/mL
Precision as CV (%)
4.8
6.3
for lamotrigine, 0.75 µg/mL for oxcarbazepine,
0.5 µg/mL for 10-hydroxycarbazepine, and 0.5
Zonisamide
µg/mL for zonisamide.
Intraday (n = 5)
Thirty-eight de-identified patient samples
Nominal concentration
5
20
30
testing positive for lamotrigine were sent to a
Mean
4.86 ± 0.37
20.24 ± 0.14
31.0 ± 2.5
Accuracy as RD (%)
2.9
1.2
3.2
nationally recognized reference laboratory for
Precision as CV (%)
7.6
0.7
8.2
comparison studies. The linear regression equaInterday (n = 12)
tion for correlation, with y as the HPLC method
Nominal concentration
20
30
described here is y = 0.98x + 0.12, with r = 0.99
Mean
20.69 ± 1.6
31.15 ± 2.4
and a standard error value of 0.61. Twenty deAccuracy as RD (%)
3.3
3.8
identified patient samples testing positive for 10Precision as CV (%)
7.9
7.8
hydroxycarbazepine were sent to a nationally
recognized reference laboratory for comparison
Oxcarbazepine
studies. The linear regression equation for correIntraday (n = 5)
lation, with y as the HPLC method described
Nominal concentration
10
20
30
here is y = 0.97x + 1.22, with r = 0.98 and a stanMean
9.93 ± 0.58
20.46 ± 1.28
30.72 ± 1.70
Accuracy as RD (%)
0.7
1.3
2.3
dard error value of 1.44. Twenty de-identified
Precision as CV (%)
5.9
6.3
5.5
patient samples testing positive for zonisamide
Interday (n = 10)
were sent to a nationally recognized reference
Nominal concentration
10
20
laboratory for comparison studies. The linear
Mean
10.26 ± 0.76
20.19 ± 0.99
regression equation for correlation, with y as the
Accuracy as RD (%)
2.6
1.0
HPLC
method described here is y = 1.0x + 0.86,
Precision as CV (%)
7.4
4.9
with r = 0.99 and a standard error value of 1.88.
10-Hydroxycarbazepine
Intraday (n = 5)
Nominal concentration
Mean
Accuracy as RD (%)
Precision as CV (%)
Interday (n = 10)
Nominal concentration
Mean
Accuracy as RD (%)
Precision as CV (%)
GC assay for levetiracetam
5
4.70 ± 0.41
6.4
8.9
* Values are reported as mean ± SD.
620
15
14.83 ± 0.43
1.1
2.9
30
31.12 ± 1.69
3.6
5.5
15
15.18 ± 0.76
1.2
5.1
30
29.92 ± 1.49
0.3
5.0
Under the described GC conditions, the retention times of levetiracetam and mepivacaine
(internal standard) were 3.8 and 6.4 min, respectively (Figure 4, patient sample). Evaluation of a
number of different extraction methods demonstrated that a simple extraction with methylene
chloride provided excellent recovery and specificity for levetiracetam. Other studies that have
used fast precipitation methods suffer from lost
of specificity and lower sensitivity (15). The
methylene chloride extraction presented here is
Journal of Chromatographic Science, Vol. 45, October 2007
Table II. Intra- and Interday Accuracy and Precision For the GC Assay for
Detetermination of Levetiracetam Concentrations in Human Plasma/Serum
(All Concentrations in µg/mL)
Levetiracetam
Intraday (n = 5)
Nominal concentration
Mean
Accuracy as RD (%)
Precision as CV (%)
Interday (n = 10)
Nominal concentration
Mean
Accuracy as RD (%)
Precision as CV (%)
5
4.74 + 0.28*
5.5
5.9
12.5
12.37 ± 0.73
1.0
5.9
25
23.52 ± 1.47
6.3
6.3
12.5
12.25 ± 0.92
2.0
7.5
25
25.47 ± 1.66
1.9
6.5
* Values are reported as mean ± SD.
equivalent in amount of effort to published solid-phase extraction methods (14,15) but likely cheaper in cost. The run-times
are faster or similar to other published methods for levetiracetam (14–17,19). A wide variety of therapeutic drugs were tested
for interference, including other anti-epileptic and neurologic
medications (and in some cases their metabolites) such as
baclofen, carbamazepine, clonazepam, epoxycarbamazepine,
ethosuximide, felbamate, gabapentin, 10-hydroxycarbazepine,
10,11-dihydroxy-trans-10,11-dihydrocarbamazepine, lamotrigine, oxcarbazepine, phenobarbital, phenytoin, primidone,
tiagabine, topiramate, valproic acid, and zonisamide. None of the
therapeutic drugs tested exhibited any interference. In addition,
no peak interferences were found in any of the batches of drugfree plasma. Twenty de-identified patient samples testing positive
for levetiracetam were sent to a nationally recognized reference
laboratory for comparison studies. The linear regression equation for correlation, with y as the HPLC method described here
is y = 0.87x + 1.47, with r = 0.96 and a standard error value of
1.59.
The method exhibited a good linearity over a concentration
range of 2.5–45 µg/mL, covering the therapeutic range. The
linear regression equation for comparison of determined versus
spiked concentration was: y = 0.99x + 0.35, with r = 0.998 and a
standard error value of 1.11. Results of intra- and interday accuracy and precision are shown in Table II. The lower limit of detection was 0.5 µg/mL and the limit of quantitation was 1.5 µg/mL.
The limits of detection and quantitation are similar to or lower
than previously reported HPLC (14,16) or GC assays (18).
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Conclusions
The HPLC method described here is simple, sensitive, and specific and allows for the simultaneous determination of three
commonly prescribed newer anti-epileptic drugs (including the
pharmacologically active metabolite of oxcarbazepine). The GC
method using a nitrogen-phosphorus detector is sensitive and
specific for levetiracetam and is suitable for clinical laboratory
use.
13.
14.
15.
Acknowledgments
M.D.K. is supported by a Clinical-Scientist
development award K08-GM074238-01A1 from
the National Institutes of Health and a
Competitive Medical Research Fund (CMRF)
grant from the University of Pittsburgh Medical
Center.
References
1. S.M. LaRoche and S.L. Helmers. The new
antiepileptic drugs: scientific review. J. Am. Med.
Assoc. 291(5): 605–14 (2004).
2. S.M. LaRoche and S.L. Helmers. The new
antiepileptic drugs: clinical applications. J. Am.
Med. Assoc. 291(5): 615–20 (2004).
M. Bialer. The pharmacokinetics and interactions of new
antiepileptic drugs: an overview. Ther. Drug Monit. 27(6): 722–26
(2005).
A.D. Fraser, W. MacNeil, A.F. Isner, and P.R. Camfield. Lamotrigine
analysis in serum by high-performance liquid chromatography.
Ther. Drug Monit. 17(2): 174–78 (1995).
G.L. Lensmeyer, B.E. Gidal, and D.A. Wiebe. Optimized high-performance liquid chromatographic method for determination of lamotrigine in serum with concomitant determination of phenytoin,
carbamazepine, and carbamazepine epoxide. Ther. Drug Monit.
19(3): 292–300 (1997).
M. Torra, M. Rodamilans, S. Arroyo, and J. Corbella. Optimized
procedure for lamotrigine analysis in serum by high-performance
liquid chromatography without interferences from other frequently
coadministered anticonvulsants. Ther. Drug Monit. 22(5): 621–25
(2000).
E. Forssblad, A.S. Eriksson, and O. Beck. Liquid chromatographic
determination of plasma lamotrigine in pediatric samples. J. Pharm.
Biomed. Anal. 14(6): 755–58 (1996).
M. Contin, M. Balboni, E. Callegati, C. Candela, F. Albani, R. Riva,
and A. Baruzzi. Simultaneous liquid chromatographic determination of lamotrigine, oxcarbazepine monohydroxy derivative and
felbamate in plasma of patients with epilepsy. J. Chromatog. B
Biomed. Sci. Appl. 828(1-2): 113–17 (2005).
M. Watelle, P. Demedts, F. Franck, P.P. De Deyn, A. Wauters, and
H. Neels. Analysis of the antiepileptic phenyltriazine compound
lamotrigine using gas chromatography with nitrogen phosphorus
detection. Ther. Drug Monit. 19(4): 460–64 (1997).
R.A. Biddlecombe, K.L. Dean, C.D. Smith, and S.C. Jeal. Validation
of a radioimmunoassay for the determination of human plasma
concentration of lamotrigine. J. Pharm. Biomed. Anal. 8(8-12):
691–94 (1990).
I.E. Leppik. The place of levetiracetam in the treatment of epilepsy.
Epilepsia. 42(Suppl 4): S44–S45 (2001).
R.A. Radtke. Pharmacokinetics of levetiracetam. Epilepsia.
42(Suppl): 24–27 (2001).
S.A. Grim, M. Ryan, M.V. Miles, P.H. Tang, R.H. Strawsburg, T.J. de
Grauw, T.A. Fakhoury, and R.J. Baumann. Correlation of levetiracetam concentrations between serum and plasma. Ther. Drug Monit.
25(1): 61–66 (2003).
J.M. Juenke, P.I. Brown, F.M. Urry, and G.A. McMillin. Drug monitoring and toxicology: a procedure for the monitoring of levetiracetam and zonisamide by HPLC-UV. J. Anal. Toxicol. 30(1): 27–30
(2006).
J. Martens-Lobenhoffer and S.M. Bode-Böger. Determination of levetiracetam in human plasma with minimal sample pretreatment.
J. Chromatog. B Biomed. Sci. Appl. 819(1): 197–200 (2005).
621
Journal of Chromatographic Science, Vol. 45, October 2007
16. B.M. Rao, R. Ravi, B.S.S. Reddy, S. Sivakumar, I.G. Chand, K.P.
Kumar, P.V.R. Acharyulu, G.O. Reddy, and M.K. Srinivasu. A validated chiral LC method for the enatioselective analysis of
Levetiracetam and its enantiomer R-α-ethyl-2-oxo-pyrrolidine
acetamide on amylose-based stationary phase. J. Pharm. Biomed.
Anal. 35(5): 1017–26 (2004).
17. V. Pucci, F. Bugamelli, R. Mandrioli, A. Ferranti, E. Kenndler, and
M.A. Raggi. High-performance liquid chromatographic determination of Levetiracetam in human plasma: comparison of different
sample clean-up procedures. Biomed. Chromatog. 18(1): 37–44
(2004).
18. M.A. Fay, R.D. Sheth, and B.E. Gidal. Oral absorption kinetics of
levetiracetam: the effect of mixing with food or enteral nutrition formulas. Clin. Ther. 27(5): 594–98 (2005).
19. N. Isoherranen, M. Roeder, S. Soback, B. Yagen, V. Schurig, and M.
Bialer. Enantioselective analysis of levetiracetam and its enantiomer
R-alpha-ethyl-2-oxo-pyrrolidine acetamide using gas chromatography and ion trap mass spectrometric detection. J. Chromatog. B
Biomed. Sci. Appl. 745(2): 325–32 (2000).
20. T. Guo, L.M. Oswald, D.R. Mendu, and S.J. Soldin. Determination
of levetiracetam in human plasma/serum/saliva by liquid chromatography-electrospray tandem mass spectrometry. Clin. Chim.
Acta. 375(1-2): 115–18 (2007).
21. G. Flesch. Overview of the clinical pharmacokinetics of oxcarbazepine. Clin. Drug Invest. 24(4): 185–203 (2004).
22. L. Franceschi and M. Furlanut. A simple method to monitor plasma
concentrations of oxcarbazepine, carbamazepine, their main
metabolites and lamotrigine in epileptic patients. Pharm. Res.
51(4): 297–302 (2005).
23. J.M. Juenke, P.I. Brown, F.M. Urry, and G.A. McMillin. Drug monitoring and toxicology: a procedure for the monitoring of oxcarbazepine metabolite by HPLC–UV. J. Chromatog. Sci. 44(1): 45-48
(2006).
24. V. Kimiskidis, M. Spanakis, I. Niopas, D. Kazis, C. Gabrieli, F.I.
Kanaze, and D. Divanoglou. Development and validation of a highperformance liquid chromatographic method for the determination
of oxcarbazepine and its main metabolites in human plasma and
cerebrospinal fluid and its applications to pharmacokinetic study.
J. Pharm. Biomed. Anal. 43(2): 763–68 (2007).
25. H. Levert, P. Odou, and H. Robert. LC determination of oxcarbazepine and its active metabolite in human serum. J. Pharm.
622
Biomed. Anal. 28(3-4): 517–25 (2002).
26. R. Mandrioli, N. Ghedini, F. Albani, E. Kenndler, and M.A. Raggi.
Liquid chromatographic determination of oxcarbazepine and its
metabolites in plasma of epileptic patients after solid-phase extraction. J. Chromatog. B Biomed. Sci. Appl. 783(1): 253–63 (2003).
27. K.M. Matar, P.J. Nicholls, M.I. al-Hassan, and A. Tekle. Rapid
micromethod for simultaneous measurement of oxcarbazepine and
its active metabolite in plasma by high-performance liquid chromatography. J. Clin. Pharm. Ther. 20(4): 229–34 (1995).
28. R. Sachdeo, A. Beydoun, S. Schachter, B. Vazquez, N. Schaul,
P. Mesenbrink, L. Kramer, and J. D’Souza. Oxcarbazepine (Trileptal)
as monotherapy in patients with partial seizures. Neurology 57(5):
864–71 (2001).
29. B. Levine, D. Green-Johnson, K.A. Moore, and D.R. Fowler.
Hydroxycarbazepine distribution in three postmortem cases.
J. Anal. Toxicol. 28(6): 509–11 (2004).
30. V. Pucci, E. Kenndler, and M.A. Raggi. Quantitation of oxcarbazepine and its metabolite in human plasma by micellar electrokinetic chromatography. Biomed. Chromatog. 17(4): 231–38
(2003).
31. M. Nakamura, K. Hirade, T. Sugiyama, and Y. Katagiri. High-performance liquid chromatographic assay of zonisamide in human
plasma using a non-porous silica column. J. Chromatog. B Biomed.
Sci. Appl. 755(1-2): 337–41 (2001).
32. R. Shimoyama, T. Ohkubo, and K. Sugawara. Monitoring of zonisamide in human breast milk and maternal plasma by solid-phase
extraction HPLC method. Biomed. Chromatog. 13(5): 370–72
(1999).
33. P. Angelis-Stoforidis, D.J. Morgan, T.J. O’Brien, and F.J.E. Vajda.
Determination of lamotrigine in human plasma by high-performance liquid chromatography. J. Chromatog. B Biomed. Sci. Appl.
727(1-2): 113–18 (1999).
34. T. Yoshida, K. Imai, S. Motohashi, S. Hamano, and M. Sato.
Simultaneous determination of zonisamide, carbamazepine and
carbamazepine-10,11-epoxide in infant serum by high-performance liquid chromatography. J. Pharm. Biomed. Anal. 41(4):
1386–90 (2006).
Manuscript received January 2, 2007;
revision received March 8, 2007.